2 The procedure
2.1.1 Barrett's oesophagus is characterised by abnormal epithelium of the oesophagus. In some patients metaplasia and dysplasia could progress to oesophageal adenocarcinoma.
2.1.2 Cancer risk is higher for patients who have Barrett's oesophagus with HGD (some of whom may already have developed early-stage cancer) than for those with LGD or no dysplasia. Patients with HGD are usually offered oesophagectomy, or frequent endoscopic surveillance and re-biopsy (to detect neoplastic changes early) followed by oesophagectomy. Endoscopic treatments aiming to remove or ablate abnormal epithelium include endoscopic mucosal resection and different ablative treatments.
2.1.3 Patients with LGD or no dysplasia are usually offered regular endoscopic surveillance and re-biopsy (with the aim of detecting potential progression to HGD or cancer).
2.2.1 PDT is carried out as an inpatient procedure, with the patient under intravenous sedation. A photosensitising agent is injected intravenously and is activated by illuminating the selected area with a laser inserted into the oesophagus. The photosensitising agent absorbs the light and forms high-energy oxygen molecules that cause necrosis of the Barrett's epithelium through a photochemical effect. For extensive Barrett's oesophagus, more than 1 treatment session may be required.
2.2.2 Patients are advised to avoid exposure to bright light and direct sunlight for several weeks after the procedure to minimise risk of photosensitivity reactions.
Sections 2.3 and 2.4 describe efficacy and safety outcomes from the published literature that the Committee considered as part of the evidence about this procedure. For more detailed information on the evidence, see the overview.
2.3.1 A randomised controlled trial (RCT) of 208 patients with HGD treated by PDT and omeprazole or omeprazole alone reported absence of HGD in 75% (104/138) and 36% (25/70) of patients respectively at 18-month follow-up (p < 0.0001) and in 48% and 4% of patients at 5-year follow-up (p < 0.0001).
2.3.2 An RCT of 72 patients without dysplasia treated by PDT or argon plasma coagulation (APC) reported complete response (reversal of columnar to squamous epithelium) in 50% (17/34) and 97% (33/34) of patients respectively at 12-month follow-up (p < 0.0001).
2.3.3 The RCT of 208 patients treated by PDT and omeprazole or omeprazole alone reported adenocarcinoma development in 15% (21/138) and 29% (20/70) of patients respectively during 5-year follow-up.
2.3.4 The Specialist Advisers listed key efficacy outcomes as reversal of dysplasia and metaplasia, and prevention of progression to adenocarcinoma.
2.4.1 One patient died 3 days after PDT treatment, with transmural oesophageal necrosis without perforation, in an RCT of 40 patients (13 with single-dose PDT vs 13 with two-dose PDT vs 14 with APC).
2.4.2 Oesophageal stricture was reported in 36% (49/138), 3% (1/34), 33% (20/60), 15% (2/13), and 27% (35/131) of patients treated by PDT in RCTs of 208, 72, 60, 26 and a non-randomised controlled trial of 199 patients respectively. Most were treated successfully with dilatation but 2 patients had perforation requiring oesophagectomy, as a result of dilatation.
2.4.3 Dysphagia was reported in 19% (absolute figures not stated) of patients treated by PDT in the RCT of 208 patients (symptom timing not stated).
2.4.4 Photosensitivity reactions within 90 days occurred in 69% of patients in the PDT arm of the RCT of 208 patients (absolute figures not stated). Photosensitivity reactions occurred in 15% (5/34) and 15% (2/13) of patients in RCTs of 72 and 26 patients respectively, and in 60% (77/129) of patients in the non-randomised controlled trial of 199 patients.
2.4.5 In the RCT of 72 patients treated by PDT or APC, buried glands were reported in 24% (4/17) and 21% (7/33) of patients respectively (difference reported as 'not-significant').
2.4.6 Specialist Advisers listed anecdotal adverse events as pain and inflammation, ulceration and severe hypotension after PDT with 5-aminolevulinic acid. They considered theoretical adverse events to include decompensation in patients with cirrhosis of the liver, and skin and retinal damage due to photosensitisation.